CA1234363A

CA1234363A - Monoclonal antibodies to surface recognition structure of mature human t lymphocyte clone

Info

Publication number: CA1234363A
Application number: CA000447354A
Authority: CA
Inventors: Ellis L. Reinherz; Stuart F. Schlossman; Stefan C. Meuer
Original assignee: Dana Farber Cancer Institute Inc
Current assignee: Dana Farber Cancer Institute Inc
Priority date: 1983-02-16
Filing date: 1984-02-14
Publication date: 1988-03-22
Also published as: ATE59411T1; EP0117114A2; US4550086A; EP0117114B1; JPH0440999B2; EP0117114A3; DE3483840D1; JPS59196822A

Abstract

Abstract of the Disclosure A monoclonal antibody which specifically binds to the surface recognition structure of a predetermined mature human T cell clone, which recognition structure renders the clone capable of acting as causative agent in a predetermined autoimmune disease, the monoclonal antibody being capable of specifically binding to the recognition structure of the clone to inhibit the ability of the clone to act as a causative agent in the predetermined autoimmune disease.

Description

ackcround of the Invention This invention relates to immunology and, more specificall~, to treating autoimmune diseases in which a mature human T cell acts as a causative agent.
A number of human autoimmune diseases, e.g.
multiple sclerosis (m.s.), sarcoidosis, juvenile type diabetes mellitus, S.L.E. (Systemic Lupus Erythmatosis), thyroiditis, rheumatoid arthritis, ankyloses spondylitis, primary biliary cirrhosis, autoimmune hemolytic anemia, immune thrombocytopenia purpura, and myesthenia gravis are caused by a subpopulation of mature T cells which deleteriously recognize and attack a specific epitope, e.g., in the case o m.s., probably myelin or Schwann cells. (As used herein, "epitope"
refers to any cell or cell product which can be recognized by a mature human T cell~) Several multiple lineage surface molecules have recently been defined on the surfaces of mature human T
cells, among them surface structures which have been designated T3, T4, and T8.- The T3 surface structure is a glycoprotein having a molecular wei~ht of about 20 kilodaltons (KD) (Reinherz et al. (1982) Cell 30, 735) t is present on the surface of all unctional mature human T celIs (i.e., it is monomorphic), and is apparently necessary for the recognition of all given epitopes by T
cells. Thus, a given monoclonal antibody (anti-T3) to T3 can block the ability of any functional T cell clone to recognize its target epitope, and contacting a heterogeneous population o mature human T cell clones , ~ ~

.

343~3 with anti-T3 wou.ld therefore block the recognition function of all of the clones ("clone", as used herein, refers to.all mature human T cells which recognize the same epitope).
We have now discovered that each different mature T cell clone has on its surface a different proteinaceous structure (hereinafter its "recognition structure") which is responsible for that clone's ability to recognize its particular target epitope. The 10 recognition structure of any given T cell clone is chemically different from the recognition structure of all other T cell clones, and recognizes a different specific epltope. For example, the T cell clone which acts as a causative agent in m.s. by attacking a 15 particular epitope (probably myelin or Schwann cells) has a recognition structure which is chemically and immunologically different from that of the T cell clone which acts as a causative agent in sarcoidosis, and this difference is the basis for the ability of these 20 different clones to recognize and attack different epitopes.
The recognition structure of each clone, ; although different fro~ all-others, has in common with that of other clones thé characteristics of l) .being 25 associated in the T cell membrane with the T3 slycoprotein, and of 2) being a heterodimer composed of a first proteinaceous compQnent having a molecular weight`of about 49 kilodaltons ("KD"), and a second : proteinaceous component having a molecular weight of 30 about 43KD, the two components being covalently linked to each other via one or more disulfide bonds.
The present invention provides a monoclonal antibody which specifically binds to the~surface ' .
:

'' recognition structure of any predetermined mature human T
cell clone, which recognition structure renders the clone capable of acting as a causative agent in a predetermined autoimmune disease. The antibody does not bind to the surface recognition structure of any other mature human T
cell clone. (For any given T cell clone, there is a potentially large number of such specifically binding monoclonal antibodies, all of which are subsumed under the singular "monoclonal antibody.") The monoclonal antibody can render inactive the predetermined mature human T cell clone by specifically binding to the clone to inhibit its ability to act as a causative agent in the predetermined autoimmune disease.
Such binding can either help cause the destruction of the clone, or can "blind" the clone, to render it unable to recognize and attack its target epitope.
The present invention provides a method of producing a monoclonal antibody which specifically binds to the surface recognition structure of a mature human T
lymphocyte clone, said T lymphocyte clone deleteriously recognizing a specific target epitope and thereby acting as a causative agent in an autoimmune disease, said method comprising providing a heterogeneous population of hybridoma cells, each producing a-monoclonal antibody reactive with said clone, selecting from said population a hybridoma producing a monoclonal antibody specifically binding to said surface recognition structure, and ~:343S3 culturing said selected hybridoma to produce said monoclonal antibody.
It also provides a method in which said surface recognition structure comprises a glycoprotein heterodimer responsible for said clone's ability to recognize said epitope.
It also provides a method in which the binding of said antibody to said surface recognition structure inhibits the ability of said clone to act as a causative agent in an autoimmune disease as well as one in which said heterodimer comprises a first component having a molecular weight of about 49 kilodaltons covalently linked via disulfide bonds to a second component having a molecular weight of about 43 kilodaltons.
It also provides a method in which said heterodimer is capable of coprecipitating with the T3 surface structure glycoprotein having a molecular weight of about 20 kilodaltons when said T3 surface structure glycoprotein is immunoprecipitated with a monoclonal antibody specific for said T3 surface structure glycoprotein.
It provides a method of providing a monoclonal antibody as set forth above including the additional step of chemically linking said antibody to a cytotoxic agent.
The invention provides a monoclonal antibody whenever made in accordance with the foregoing methods.

'~?"'~ - 3a -~3~3Çi3 The invention also provides a method of inhibiting the capacity of a mature human T cell clone to recognize the epitope for which its surface recognition structure is specific comprising contacting said T cell clone with a monoclonal antibody which is specific to said surface structure, in an amount sufficient to cause said inhibition.
In more detail, the steps involved in obtaining the desired specific monoclonal antibody are as follows.
First there is provided a mature human T cell clone which is to be inactivated, e.g. the T cell clone involved in multiple sclerosis. The T cell clone is then used to immunize a mammal, e.g. a mouse. Appropriate lymphocytes (usually spleen cells) from the 3b ~23~L3~:;3 mammal are then harvested and fused to appropriate transformed cells (usually mouse or human myeloma cells), forming a heterogeneous population of hybridomas. The hybridoma population is then screened for proàuction of monoclonal antibody, and an.ibody-producing hybridomas are then screened for those producing monoclonal antibody specific to the recognition struct~re of the immunizing T cell clone.
This second screening step can be accomplished by contacting the monoclonal antibodies produced by the differen~ hybridomas with a plurality of different mature human T cell clones, including the T cell clone used for immunization, and selecting the clone or clones reactive only with the immunizing clone, and discarding those reactive with any of the other clones (these proa~ce antibodies which react with a st~ucture, e.g.
T3, on the immunizing clone which is found on the sur~ace of other clones in addition to the immunizing clone). Enough different T cell clones must be used for screening to insure specificity; generally at least 10, and preferably 20 or more, clones should be used. The selected hybridoma is then cultured to produce the desired specific monoc~onal antibody.
The method by which the immunizing T cell clone (the clone involved in the autoimmune disease) is obtained will depend on the nature of the disease, e.g.
the primary anatomical location where the deleterious attacking of the epitope by the T cell clone occurs.
For example, since m.s. is primarily a disease of the brain, and the brain is in contact with the circulating cerebrospinal fluid, the T cell clone of interest will be present in that fluid. Furthermore, in most .

.

3L~;3 ~36~

instances, the only activated T cell clone present in the fluid of an m.s. patient will be the T cell clone involved in the disease. Thus, to obtain the attacking T cell clone from an m.s. patient, cerebrospinal fluid is withdra~n, e.g. by a simple lumbar punct~re, and the T cell population is expanded, using conventional techniques, with Interleukin-2 ("IL-2"), whlch induces proliferation only in activated, and not resting, T cell clones, since only activated clones possess IL-2 receptors.
Another example of an anatomically limited autoimmune disease is sarcoidosis, in which an activated T4 cLone attacks the lungs. To obtain the attacking clone, fluid can be obtained from the bronchial tree and I~-2 expansion carried out to proliferate the activated attac~ing clone.
In the case of an a~toimmune disease in which tne attacking T cell clone is present in the blood, which contains a huge number of mature T cell clones, IL-2 is also used, since almost all of those circulating clones will be resting clones which are not stimulated by IL-2.
As is clear fEom the above, in most cases it is not necessary to identify the epitope being attacked in order to isolate the attacking clone; in many cases the only activated T cell clone, or at least the major activated clone (the clone present in the greatest n~mber) will be the attacking clone, which can be isolated without determining precisely what epitope it recognizes.
The monoclonal antibody produced by the above method can be contacted with the delet-rious T cell .

.
.

3~3~;3 clone to help destroy it or to blind it to inhibit its capac-ty to recognize and attack the target epitope it recognizes. The monoclonal antibody is thus useful in th~ treatment of an autoimmune disease, e.g. S.L.E., rheumatoid arthritis, or multiple sclerosis, in which a mature T cell clone to which the monoclonal antibody is specific acts as a causative agent by recognizing and attacking a particular epitope.
~hen administered to a patient having the autoimmune disease involving the T cell clone to which the monoclonal antibody is specific, the monoclonal an.ibody selectively binds to the clonal eells, either helping destroy them or blinding them so that they are unable to recognize and attack their target epitope, thus improving the medical condition of the patient.
The specificity of the monoclonal antibodies of the invention means that administration inactivates only the deleterious T eell elone, and has no effect on the patient's remaining mature T cells, whieh make up a crucial portion of the immune system. Furthermore, the fact that such specificity is a function of the particular epitope recognized by the recognition structure means that a particular monoelonal antibody will be effective against all patients having an autoimmune disease in whieh a mature T eell elone attacks the same epitope. The invention also permits the tailoring of a monoclonal antibody to any predetermined mature T cell clone of a given patient.
Any of the antibodies of the invention ean be chemically linked to a eytotoxie agent to seleetively deliver the toxin to the predetermined clone, without harming other cell types. Such cytotoxic agents can ~.X3~363 incl~de chemothera~eutic agents, biologieal to.cins such as ricin and mushroom toxins, radioaetive agents, and photoactive toxins which are activated by W light.
Alternatively, the antibody can be used not to destroy but to blind the predetermined clone, in which ease no cytotoxic agent is employed.
The method of administration of a monoclonal antibody of the invention will vary with individual circ~mstanees, e.g. the partieular disease being treated, as will the dosage and frequeney of administrationO Generall~y, the antibody will be mixed, prior to administration, with a non-toxic, pharmaceutically acceptable earrier substance, e.g.
normal saline, and ~ill be administered using any lS meàicall~ appropriate procedure, e.g., intravenous administration. The anti~ody will generally be present in the carrier in a eoncentration of between about 0.5 ~g antibody/ml and S00 ~g antibody/ml. The amount of antibody administered at one time ~ill generally range between about 50 ~g and 500 ~g of antibody per kg of body weight. It will be desirable in some instanees to administer the antibody to the patient in a series of more than one administration, and regular periodie administration will sometimes be re~uired.
The following speeifie example is intended t~
more partieularly point out the invention, without aeting as a limitation upon its seope.
xample A monoelonal antibody speeifie to the surfaee reeognition strueture of a eytotoxie mature human T eell elone, designated CT8III, was produeed aeeording to the following proeedure.

~L2343~;3 First, the cytotoxic CT8III clone was generated from a human subject's mononuclear ceLls, which ~lere stimulated in MLC witn allogeneic ~pstein Earr '~irus (EBV) - transformed lymphoblastoid cells, of 5 the ceIl line designated Laz 156. Cloning and recloning ~ere carried out to propogate CT8III, according to the methoa described in ~;leuer et al. (1982) P.N.A.S. USA 7~, ~590. Optimal growth of T cell clones was obtained in tne presence of a feeder cell suspension of irradiated autologous'peripheral blood mononuclear eells plus irradiated Laz 156 cells in IL-2 conditioned media;
prolireration also occurred in the presence of either irradiated Laz 156 or IL-2 conditioned medium.
The pheno-type of CT8III, as determined with a panel of monoclonal antibodies by means of indireet i~munofluorescence, was Tl-,T3+,T4-,T6-,T8+,Tll+, T12~,Ia~. (+ means reaetive, - means unreactive). The target specificity, as analyzed on a panel of HLA typed E~V-transformed target cells, was HLA-A3. The lympholytic function of CT8III could be blocked by anti-~L~ but not by anti-Ia antibodies on the target cell level.
Balb/cJ miee were initially immunized IP with 5 ~ 10 CT8III cells in PBS. Ten days later a booster injection with 5 x 106 CT8III cells IP was per~ormed. Three days prior to somatie fusion of immune splenocytes with NS-l myeloma cells, two mice were in jected IV and IP with a total of 5 x 106 CT8 cells. After sacrific of the animals, immune splenocytes were obtained and fusion performed as described in Reinherz et al. (1979) J. Immunol. 123, 1312. hybridoma growth was evident by weeks.

.

j: :
:
:

~3~6~

Subsequently, supernatants of hybridoma clones were screened for reactivity on CT8III by means of indirect immunofl~orescence. Of 221 individual hybridomas which ~lere estaDlished in HAT medium, 5 secreted monoclonal antibodies reactive with the immunizing clone CT8III.
Those which failed to produce antibodies reactive with CT8III were discarded.
Reactive supernatants were then screened on a large panel of cell types. These included: antologous and allogeneic resting peripheral blood cells (T cells clones, ~ cells, macrophages, granulocytes, platelets);
thvmocytes; autologous and allogeneic activated T cells (~IL~ lymphoblasts, Con A lymphoblasts, T cell lines (2-~months old), 7g additional T cell clones derived from the same donor at CT8III, autologous and allogeneic B-cell lines (Laz 503, Laz 156, Laz 471), T cell tumbr lines (CE~1, Molt 4, HSB~; and tumor lines of non-T cell lineage (K562,HL-60, Laz 221, KGl). It was found that 43 of the 59 CT8III-reactive antibodies were also reactive with the autologous B lymphoblastoid cell line, Laz 509, and were therefore most likely directed at either alloantigens or other broadly distributed cell surface molecules. In contrast, the remaining 16 antiboàies were reactive with the immunizing clone but lacked reactivity with ~ cells, macrophages, granulocytes, platelets, B lymphoblastoid lines, myeloid lines, and other hematopoietic lines.
Within the latter group, 7 antibodies were reactive with the 76KD T~ surface structure (anti-T8D J), one recognized the 20KD T3 molecule (anti-T33), and 6 were specific for activated but not resting T lymphocytes (anti-TAl ~). In addition, two , , .

~3~363 antiboaies, termed anti-TilA and anti-TilB, reacted e~cl~slvely with the CT8III clone and not with human t~mocytes, with resting or activated human peri?heral T
cells (~oth autologous and allogenic) or with the 7~
addi~ional individual alloreactive T cell clones on the panel. Anti-TilA and anti-Til~ were of the IgGl and Ig~l isoty2e, respectively. The hybridoma line producing anti Til~ was deposited in the American Ty~e Culture Collection, Roc~ville, ~laryland, on February 15, 1983, and was given ATCC Accession ~umber HB 8213.
The specificity of anti-Til and anti TilB
was in contrast to the reactivity of anti-T3~ whicn reacts with al mature T cell clones.
Further studies were performed on the speci-ic anti-TilA and anti-TilE antioodies, which were proviaed in quantity needed for tests by cloning the hybrid cells producing them by limited dilution, and then injecting individual hybrid cells into pristine, primed ~alb/cJ mice. The resulting ascitic fluid was used as a source of antibody.
In an antibody binding study employing CT8III
and another mature T cell clone (designated CT8IV) not reactive with anti-TilA or anti-TilB, and from the same donor as CT8III, all CT8III cells were:shown to 25 be reactive with anti-TilA and anti-TilB, whereas CT3IV cells are unreactive with either antibody, as analyzed by indirect immunofluorescence on an Epics V
cell sor.er. In contrast, both CT8III and CT8Iv expressed the 20~D T3 molecule as defined by anti-T3 reactivity. Similar results to those obtained with CT8IV were found for the other 79 autologous clones.
Reactivity was determined by means of individual .:.!

, ., 3~3~3 immunorluorescence assays with goat anti-mouse F(ab')2-FITC.
The following study demonstrated that anti-Ti1A and anti-Til~ defined a surface structure on CT8III involved in epitope (antigen) recognition, and that these antibodies were capable or blocking such recognition. To study cell-mediated lympholysis and t~e effects of the antibodies on lympholysis, CT8III cells were incubated for various periods (30 min. to 18 hours) with anti-TilA and anti-TilB at several dilutions, prior to addition of ~lCr-labeled target cells. In parallel, the effects on lympholysis of several other T
cell-specific monoclonal entibodies (anti-T3A, -T3B, -T8A, -T8C, and -T12) were investigated. The study was performed at an E/T ratio of 20:1 in V bottom microtiter plates (Falcon, O~nard, CA) according to the method described in Reinherz et al. (1979) P.N.A.S. USA
76, 40610 CT8III cells, in the absence of anti-TilA
or TilB, were found to be highly ef icient (greater than 50% lysis) in lysing the target cells, the HLA A 3+
; B lymphoblastoid line Laz 156. The two specific antibodies Anti-TilA and -TilB both markedly inhibited CT8III's lympholytic capacity (less than 20%
and less than 5% lysis, respectively). These inhibitor effects were observed at dilutions of ascites varying rom 1:250 to 1:2500 and were evident after less than 30 min. of preincubation with CT8III. The magnitude of inhibitory effect increased with the duration of preincubation. (Anti-T3A B and anti-T8A alsoi not surprisingly, inhibited the lympholytic effect of CT8III ' ) ; .

,: .

~;23g~ii3 In another lympholysis study designed to corroborate the indirect immunofluorescence results demonstrating that anti-Til A and -Ti,~ were exclusively reactive with CT8,ll, the inhibitor effects of these antibodies were tested against ~ lympholytic clones in addition to CT8lll. As shown in Table 1, below, these two specific antibodies were effective in inhibiting cell-mediated lympholysis only against CT8lll, and not against the other five clonesO In contrast, the effects of anti-T3, anti-T4 and anti-T8 antibodies were not restricted to the CT81ll clone.
Anti-T3 monoclonal antibodies blocked all 5 lympholytic clones regardless of their T4+ or T8+ subset derivation or specificity; anti-T4A inhibited cytolysis by the T4+
: clones CT41 and CT41, and had no effect on killing by the T8+ clones CT8l, CT8ll and CT8lll; and in a reciprocal fashion, anti-T8A and anti-T8~ monoclonal anibodies inhibited killing of all three T8+ clones but not the two T4 clones, CT4, and CT4ll.

- 12a - ~IL2343Ç~3 412-1417 .~ _ _ o ~ 1+~

~ r-- ~ ¦ +
C~ ~_ o ~1 ~ O
L~ , ~r_ ~~ r- ~> C~` 00 r~ r-~~ O r~ ~ r~ ~ r- ~ r r~
C ~ C C C C

~239L~363 Because some previous studies have indicated that some lympholytic cèlls not only specifically kill target cells but also proliferate to them in an antigen-specific fashion, the following prolirerative S stud~ was perfor;ned to examine the ef ect of anti-TilA
and-Til2 on the antigen-specific proliferative ~ 8III. CT8III cells were incubated ~or varying periods ~30 min. to 18 hours) with the antibodies at several dilutions. Subsequently, untreated or antikod~ treated cells were plated, at 15,000 cells/well, into round-bottom microtiter plates ~Costar, Cambridge, MA) along with eitner medium ~RPMI
16~0 supplemented with 10% human AB serum); irradiated Laz 156 cells (15,000 cells/well); IL-2 conditioned medium ~final concentration 5%); or Laz 156 cells plus IL-2. ~IL-2 containing supernatants were produced by stimulating whole peripheral blood mononuclear cells with ph~Jtohemagglutinin ~PHA) and phorbol myristate (P~IA) acetate in the presence of irradiated Laz 156 cells, as described in ~euer et al~ ~1982) P.N.A.S. ~SA
79, ~59C.) Following a 24 hour incubation at 37C, the various cultures were pulsed with 0.2 Ci of tritiated thymidine (Schwartz Mann, Division of Becton Dickinson, Orangeburg, NY) and harvested 18 hours later on a Mash II apparatus (Microbiological Associates, Bethesda, MD)- H-TdR incorporation was then measured in a Packard scintillation counter (Packard Instrument Co., Downer's Grove, IL).
The results of the proliferation study are shown in Table II below (each value represents the lnean of triplicates). As shown therein, untreated CT8 .
' ~3~L~63 cells ?roliferated to IL-2 containing supernatants as well as to the allogeneic cell line, Laz 156 to which they had been originally stim~lated. Under these e~?erimental conditions, antigen (Laz 156) was the st~onger stimul~s Eor CT8III cell proli~eration, as juged by the greater H-TdR incorporation into CTBIII with the former ( 12,000 cpm vs. 2,71~ cpm).
In addi~ion, the induction of clonal proliferation by the combination of alloantigen and IL-2 was greater than with either alone (17,152 cpm). Pretreatment of CT3III with anti-TilA or anti-Til~ markedly reduced antigen specific proliferation of the CT8 clone. Thus, whereas the untreated CT8III clone proiiferated ~ith greater than 12,000 counts of 3H-TdR
to Laz 156, CT8III preincubated with anti-TilA or anti-Tilp proliferated with less than 2,500 counts/ a reduction in proliferation of greater than 80~. A
similar reduction of proliferation was obtained with anti-T3B. That these effects were not simply due to an inactivation of the CT8III clone as a result of antibody treatment is clear from the facts that 1) anti-T12 and anti-T8A had no inhibitory effects in this system (Table 2); and 2) the anti~Til~, anti-TilB or anti-T3B treated CT8III clone had an augmented proliferative capacity to IL-2 containing supernatant (greater than 10,000 cpm vs less than 3,000 cpm). In contrast, anti-T8A has the ability to profo~ndly block the lympholytic effect of T8+
lympholytic clones, and completely lacks any inhibitory effect on antigen specific proliferation.

.

~;~3~3~;3 o U~ ~ o ,_ tn ~. 1 + + +
~> ~ ~ cr~
C~J o C
s C ~ ", _ ,_ ~c; I N
Lr~
o c ~ ~ m ~
J ~
1 0 o C o o~
V~
Z ~ +~
C _ ~ r~
. ~ ~,C .
LL , Y ~ ~ ~ o ~
~ ¦ + ~ N
C
O

2 0 ", ~D o c: u ~ C~J ~ o _~ ~ , ~, +, +, 1 1 , ~: r5~ ~ - C~
_ O C~l ~_ C C
O
~ ~ 25 o ~ ~oC~ +, ~, +, ~, 1~ L
C N . ~.J 1~
ri _t ,~
~n :J
: ~ E e ~ _ ~ : ' ~Z3~ Ei3 T~e observation that anti-T3, anti-Til~, and anti-Til~ all inhibited antigen specific pro?liferation, lympholytic function, and enhanced IL-2 res~onsiveness of CT8III, suggested that there might - 5 be a relationship between the cell surface structures defined by these antibodies. In this regard, previous st~dies indicated that binding of anti-T3 antibodies to T cells at 37C resulted in selective modulation and e:cternal shedding of the T3 molecular complex without affecting cell viability or altering expression of other known T cell surface structures including Tl, Tll, T12, or T8 surface molecules. To determine whether anti-T3 induced modulation produced changes in surface ex?ression of detectaole TilA or TilB molecules, CT8III cells were first incubated in final culture medium RPI~I 1640 plus 12% human AB serum-with anti-T3B
for 18 hours at 37C, then washed to remove free monoclonal antibody, and subsequently, cell reactivity analyzed by indirect immunofluorescence on a Epics V
fluorescence activated cell sorter with a panel of monoclonal antibodies.
Prior to modulation, all CT8III cells were found to be reactive w~h anti-T3A, anti-TilA, anti-TilB and anti-T8A. In contrast, after modulation with anti-T3B, T3 antigen was no longer detectable. Anti-T3B induced moduation also resulted in loss of the anti-TilA and anti-TilB surface epitopes. That this was not a nonspecific effect is evident from the observation that the T8 antigen density was uninfluenced by this process.
Incubation of CT8III cells with either anti-TilA or anti-TilB had identical effects: in ~3~3~3 all cases T3, TilA and TilB molecules comodulated.
These results indicate that the molecules defined by anti-T3, antl-TilA and anti-Til~ are functionally and phenotypically linked on the cell surface of the CT8III clone-To investigate the relationship o~ the epitopede~ined by anti-T3 and those defined by anti-TilA and anti-TilB further, both competitive antibody binding inhibition studies and immunoprecipitations were performed.
For these studies, directly FITC labelled purified monoclonal antibodies (anti-T3A, anti-TilA, anti-TilB) were prepared. All studies were perfomed at 4C in order to prevent modulation of cell surface antigens. In the first incubation step, CT8III cells were incubated with saturatlng amounts of one or another unlabelled monoclonal antibody for 30 min. Then the cells were washed twice and incubated with directly FITC
labelled monoclonal antibody (30 min) prior to FACS
analysis~ Unlabelled anti-T3A and anti-T3B
inhibited subsequent binding of directly FITC labelled anti-T3A, whereas anti-TilA and anti-TilB did not. In a reciprocal fashion, unlabelled anti-T3A or anti-T3B failed to inhibit subsequent binding of directly FITC labelled anti-TilA or anti-TilB. In contrast, either unlabelled anti-Til antibody was inhibitory for FITC labelled anti-TilA or anti-TilB. Under no circumstances did unlabelled anti-T8A block binding of directly FITC labelled anti-T3A, anti-TilA, or anti-TilB. Taken together, these findings are evidence that there is a single surface epitope recognized by both anti-TilA
and anti-TilB.

~. .

~3~ Eii3 To biochemicall~ define the surCace structJre detected by anti-TilA and anti-TilB, CT8III cells were labelled by lactopero;;idase technique with (~el England Nuclear, ~oston, MA) and immunoprecipitates from soluhilized memhranes subjected to SDS
polvacr~flamide gel electrophoresis, as follows. To 20 loG cells sus?ended in 1 ml PBS ~ere added successively 10 ~1 glucose (0.5 Mol/l), 5 ~1 NaI t5 :c L~lol/l), 10 ~1 lactoperoxidase (2 mg/ml), 1 mCi Na 131I and 20 ~1 glucose oxidase (7.5 mU/ml). This mixture was incubated for 15 min at room temperature followed by addition of 100 ~1 NaI (1 Mol/l). After an additional 2 min of incubation, cells were washed 4 times in Hanks balanced salt solution. The final pellet was l~sed in 500 ~ 5 diluted RIPA stock solution containing 1o Triton ~-100 (RIPA stock solution: 0.1 ~IOl~l Na~2PO~, lm Mol/l P~lSF, 10 m Mol/l EDTA, 10 m Mol/1 EGTA, 10 mMol NaF, 1% deoxycholate sodium salt, aprotinin 200 KIU/ml, pH 7.2~. The suspension was centrifuged for 5 min at 1120g and the resulting supernatant precleared twice utilizing monoclonal antibody anti-T6 covalently linked to CnBr activated Sepharose~ B (Pharmacia, uppsala, Sweden) each for 30 min at 4C and subsequently centrifuged 5 mln at 1120g.
Precleared lysates were incubated with monoclonal antibodies bound to Cn~r activated Sepharose~4~ for 60 min at 4C. The resulting precipitate was subsequently washed 5 times in RIPA solution (stock 1:10 diluted) +
1% Triton X-100 resuspended in gel buffer (0.125 ~ol/l Tris-HCl pH 6.8 containing 106 glycerol, 3% SDS and 56 2-mercaptoethanol) and boiled for 5 min. SDS-PAGE was performed on a continuous vertical slab gel (12~5Po c~ k ~L~3~363 polyacrylamide) for 14 hours according to a modification of the procedure described in Laemmli (1970) Nature 227, 680.
As a control, CT8IV (unreactive with anti-TilA and anti-TilB) was tested in parallel.
Molecular weights were estimated from the mobility of radioactive molecular weight markers (New England Nuclear) (Cytochrome C, 12,300; lactoglobulin A, 18,367;
carbonic anhydrase, 30,000; ovalbumin, 46,000;
phosphorylase AB, 97,400). Under reducing conditions, anti-TilA precipitated two distinct bands of approximately 49KD and 43KD, respectively from the CT8III clone, but not the irrelevant CTaIV clone.
In a parallel fashion, anti-TilB precipitated identical structures from the same CT8III clone but not the irrelevant CT8IV clone. Both the inability of anti-TilA and -TilB to inhibit anti-T3 cell surface binding and the present biochemical data are consistent with the fact that these two specific antibodies are directed at a different glycoprotein than the major 20KD
glycoprotein defined by anti-T3 antibodies.
To determine whether the 49KD and 43KD
components of the CT8III surface recognition structure to which anti-TilB specifically reacted were disulfide linked on the mem~rane of CT8III, the anti-TilB
precipitates were run in SDS-PAGE under non-reducing conditions, i.e., 2-mercaptoethanol was omitted. Under such condi~ions, anti-TilB immunoprecipated a single broad band at 80-9OKD, indicating that the surface recognition structure is a heterodimer composed of subunits of molecular weight 49KD and 43KD, joined by disulfide bonds.

~3~3G3 As was mentioned above, the T3 structure and the recognltion structure are associated on CT8~
To define the relationship of these structures urther, anti-T3 monoclonal antibodv was used to immunopreripitate surface material from CT3III and 6 additio~al 125I labelled antigen responsive clones wi.h difCering specificities which were derived from the same donor. The precipitations were characterized on SDS-?AG~ analysis and compared to those obtained with 10 anti-Til~.
The results obtained for CT8III and CT8IV
sh~wed that the material precipitated with anti-T3 from the two clones appeared to be similar, the major protein band having a molecular weight of abou~ 20~D. In both cases there were also bands defining proteins of molecular weights of 49~D and 43KD. These appear to be identical in size to the protein found in the anti-TilB immunoprecipitate from CT8III. As expected, anti-TilB did not precipitate any detecta~le banas from the TilB unreactive clone CT8I~.
To determine whether the higher molecular weight proteins in the anti-T3 immunoprecipitates from CT8III were, in fact, rPlated in those defined by the anti-clonotypic antibody anti-TilB, a series of sequential immunoprecipitation studies were performed with the TilB unreactive clone CT8II and the TilB
reactive cl e CT8III (CT3II was a mature lympholytic T cell taken from the same individual as CT8III.) For the sequential precipitation studies, radiolabelled cell lysates were precleared 2 times with anti-T12-Sepharose or anti-Til~-Sepharose, incubated at 4C for 60 min with anti-T3, and precipitated with protein A Sepharose-rabbit antimouse Ig.

~3~3$3 Preclering of externally labelled cell lysates from CT~II with anti-T12 or anti-TilB did not affect the subsequent density oE the higher molecular weight bands in the anti-T3 immunoprecipitation on SDS-P~GE
anal~sis. In cont!ast, when labelled cell lysates from the Til~ reactive clone CT3III were precleared with anti-Til~ prior to anti-T3 immuncprecipitation, there was a mar~ed reduction in the hig~er molecular weight bands in comparison to the lysates of CT8III
precleared with the unrelated anti-T12 antibody. Since anti-Til~ precipitated little, if any, material in the range of the 20KD T3 molecule, it is likely that the monoclonal antibody dissociates the 49/43KD molecule from the 20/25KD molecules when it binds to the former.
The above results provide strong evidence that the 49KD
and ~3KD proteins found in anti-T3 immuno~recipitates of CT8III cell lysates comprise the recognition structure recognized by anti-TilB. All of the results also indicate that clonotypic structures exist on all mature T cell clones and are associated with the 20 and 25KD T3 molecules. The latter findings suggest that anti-T3 could be used as a probe to isolate and compare these clonotypic structures on a series of other clones where no anti-clonotypic antibodies reactive d~irectly with the 49/~3KD structure yet exists.
; The finding that Ti1A and TilB were present on CT3III and none of 79 additional clones from the same individual is evidence that they define a structure with a variable domain. To determine whether the 4~/43KD structrue on clones of different specificities do, in fact, difEer from one another, 2D gel electrophoresis and peptide map analysis of anti-T3

3~L3~3 immunoprecipitates from multiple e~ternally labelled clones were compared. The labeled clones were CT8II
and C~3III; two autoreacti-Je lympholytic clones AT4I
and (Arr~ ); and 5 aàditional T4+ and T8+ T cell clones.
For 2 dimensional gel electrophoresis, anti-T3 preci?itates from lysates of the above 1 5I labelled clones were dissolved in isoelectric focusing sample buf~er (9.5 M urea, 2% triton X-100, 2~ ampholytes; pH
3.5 - 10, pH 4-6, and pH 5-8; ratio 1:1.2:4) and analyzed by isoelectric focusing gels and a SDS-PAGE
system according to the methods described in Garrels (1~79) J. Biol. Chem. 254, 7961 and O'Farrel (19~5) J. ~iol. Chem. 250, 4007.
Peptide mapping by limited proteolysis employed the method described in Cleveland et al (1977) J. Biol.
Chem. 252, 1102, with a modification that the first dimensional separation was performed in 12.5~
SDS-pol~acrylamide gel tubes. After being subject to electrophoresis, the latter was incubated for 30 min at 20C with 100 mM Tris-HCl, pH 8.6 and 100 9 per ml protease ~either chymotrypsin (Boehringer Mannheim, Indianappolis, IN) or s`~aph V8 (Miles Lakoratory, ~l~hart, IN)~. These were then applied to 12.5%
polyacrvlamide gels containing an identical concentration of protease in the stac~ing gel.
In the 2D gel analyses and isoelectric focusing, the 20KD glycoprotein within the anti-T3 ~immunoprecipitate resolved into 5 spots in this system ; 30 whereas the 25KD glycoprotein appeared as at least two separate spots which were more basic than the former.
From the 2D analysis, it was concluded that these -' .

~3~3~;3 structures on AT4I and ~T4III clones are similar, if not identical. Comparable results were also obtained with the five additional T4+ and T8+ T cell clones tested. In contrast, the 4~/43KD clonotypic structures showea considerable variability. For example, the ~3KD
protein in the immunoprecipitate of AT4I had a more acidic isoelectric point than the 43KD protein from AT9III. Some charge variability was noted in the 49KD
protein rom these clones.
From the above results it can be concluded that among individual T cell clones the 20 and 25 KD proteins show little heterogeneity whereas the 49/43KD
heterodimer shows considerable isoelectri~ point variability.
lS To define whether this was due to variablility in the polypeptide chains themselves, the 49/43KD
structures within anti-T3 immunoprecipitates of individual clones were obtained, digested, and peptide maps produced. The results indicated that there is one ~ 20 major chymotryptic fragment generated from the 49KD
; structure. This appears identical for each of these clones. In contrast, chymotryptic fragments from the 43~D protein of CT4II~and ~T8III are distinct. The peptide map of CT8III shows a single major cleavage ~25 product and multiple minor fragments whereas that from ; CT4II contained a different peptide fragment. The latter ~ragment is clearly not present in the map of the 43KD protein from CT8III.
Analysis of the 49 and 43KD glycoproteins from

4 additional clones showed considerable heterogeneity in chymotriptic as well as staph V53 maps. However, while variability was always evident within the 43KD

: `

.' .
.

~23~3~;3 structure, the possibilit~y could not be ruled out that some variability might also e~ist within the 4~KD
molecule as well. In contrast, in no instance did one detect variability within the 20/25KD glycoproteins of t~e anti-~3 immunoprecipitates.
Other embodiments are ~ithin the following cl~i~s.

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method of producing a monoclonal antibody which specifically binds to the surface recognition structure of a mature human T lymphocyte clone, said structure comprising a glycoprotein heterodimer responsible for said clone's ability to deleteriously recognize a specific target epitope and thereby act as a causative agent in an autoimmune disease, said method comprising providing a heterogeneous population of hybridoma cells, each producing a monoclonal antibody reactive with said clone, selecting from said population a hybridoma producing a monoclonal antibody specifically binding to said surface recognition structure and thereby inhibiting the ability of said clone to act as a causative agent in an autoimmune disease, and culturing said selected hybridoma to produce said monoclonal antibody.

2. A method as claimed in claim 1 in which said heterodimer comprises a first component having a molecular weight of about 49 kilodaltons covalently linked via disulfide bonds to a second component having a molecular weight of about 43 kilodaltons.

3. A method as claimed in claim 2 in which said heterodimer is capable of coprecipitating with the T3 surface structure glycoprotein having a molecular weight of about 20 kilodaltons when said T3 surface structure glycoprotein is immunoprecipitated with a monoclonal antibody specific for said T3 surface structure glycoprotein.

4. A method as claimed in any of claims 1 to 3 in which said autoimmune disease is Systemic Lupus Erythmatosis.

5. A method as claimed in any of claims 1 to 3 including the additional step of chemcially linking said antibody to a cytotoxic agent.

6. A method as claimed in any of claims 1 to 3 including the additional step of chemically linking said antibody to a cytotoxic chemotherapeutic agent.

7. A method as claimed in any of claims 1 to 3 including the additional step of chemically linking said antibody to a cytotoxic photoactivated toxin.

8. A method as claimed in any of claims 1 to 3 including the additional step of chemically linking said antibody to a cytotoxic radioactive agent.

9. A monoclonal antibody which specifically binds to the surface recognition structure of a mature human T lymphocyte clone, said T lymphocyte clone deleteriously recognizing a specific target epitope and thereby acting as a causative agent in an autoimmune disease.

10. A monoclonal antibody as claimed in claim 9 in which said autoimmune disease is Systemic Lupus Erythmatosis.

11. A monoclonal antibody whenever produced by the method of any of claims 1 to 3 including the additional step of chemically linking said antibody to a cytotoxic agent.

12. A monoclonal antibody whenever produced by the method of any of claims 1 to 3 including the additional step of chemically linking said antibody to a cytotoxic photoactivated toxin.

13. A method as claimed in any of claims 1 to 3 including the additional step of chemically linking said antibody to a cytotoxic radioactive agent.